Romain == Romain Francois [EMAIL PROTECTED]
on Thu, 06 Oct 2005 23:23:12 +0200 writes:
Romain Selon bogdan romocea [EMAIL PROTECTED]:
A related comment - don't rely (too much) on boxplots. They show only
a few things, which may be limiting in many cases and completely
Hi all,
Thanks for the quick response. I see the ade4 package in not needed
for distance matrix computation, but as far i can see you need it for
comparing two distance matrices. In the stats package i can't find any
similiar functions like mantel.randtest or RVdist.randtest of the ade4
package.
Hi
it seems that the problem is why you got all your numerics
converted to factors. If taken from some spreadsheet there could
have been some unnoticed ***spaces*** in blank cells which
turned the numeric column into factor column.
HTH
Petr
On 6 Oct 2005 at 17:08, Florence Combes wrote:
hi,
but I don't see why you would need an extra package ade4 and
its extra - function mat2dist().
when the 'stats' package already provides the function
as.dist(.) {the help page of which was mentioned by the
original poster}.
it's just a matter of habit and it's just a proposition.
I
Hi
I tried to compile R-2.2.0 just now. configure worked fine, but
compilation stopped with
[snip]
src/include -I/sw/include -I/usr/local/include -DHAVE_CONFIG_H -fno-
common -g -O2 -c util.c -o util.lo
gcc -I../../src/extra/zlib -I../../src/extra/bzip2 -I../../src/extra/
pcre
Le 07.10.2005 09:21, Martin Maechler a écrit :
Romain == Romain Francois [EMAIL PROTECTED]
on Thu, 06 Oct 2005 23:23:12 +0200 writes:
Romain Selon bogdan romocea [EMAIL PROTECTED]:
A related comment - don't rely (too much) on boxplots. They show only
a few things,
On Fri, 2005-10-07 at 09:31 +0200, Mattias de Hollander wrote:
Hi all,
Thanks for the quick response. I see the ade4 package in not needed
for distance matrix computation, but as far i can see you need it for
comparing two distance matrices. In the stats package i can't find any
similiar
Hi,
The question is as follows: has anyone coded panel data unit root tests
with R? Even the first generation tests (see e.g. Levin Lin 1993;
Pesaran, Smith Im 1996; Maddala Wu 1999) would be sufficient for my
needs. To my understanding, these are rather easy to code, but as I have
taken
Take this as an example:
a=data.frame(col1=c(1,2,3,4,5), col2=c
(my,beloved,daughter,son,wife))
b=data.frame(col1=c(1,2,4),
col2=c(my,beloved,son))
a
col1 col2
11 my
2
2 beloved
33 daughter
44 son
55 wife
b
col1col2
11 my
22
try this (presuming that you want to find the missing lines based on
the 2nd column of the data frames):
a[!a$col2 %in% b$col2, ]
I hope it helps.
Best,
Dimitris
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address:
On Fri, 7 Oct 2005, Robin Hankin wrote:
Hi
I tried to compile R-2.2.0 just now.
On MacOS X.
configure worked fine, but
compilation stopped with
Please ask MacOS-specific questions on the r-sig-mac list.
(Missing RestFP, SaveFP.)
I believe this is well-known and indicates an incompatible
Vittorio,
you can also code sthing like this:
ind-(setdiff(b,a))$col1
you have what is common in a and b,
and then you take the opposite in a :
a[-ind,]
col1 col2
3 3 daughter
5 5 wife
hope this helps,
florence.
On 10/7/05, Vittorio [EMAIL PROTECTED] wrote:
Take this as an example:
It's probably a beginner's question:
How do I show text in boxes?
That is, can I specify a background color for text output with text() ?
The following doesn't work as I would expect:
text(labels=123, 50, 0.5, bg=green)
I've experimented with legend(),which will make the box too wide, and
As an alternative to the effects package, try predict() with
type=terms
JM
On 7 Oct 2005, at 8:00 PM, Peter Dunn wrote:
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Peter Dunn
Sent: Wednesday, October 05, 2005 9:06 PM
To: R-help mailing list
Subject: [R] R/S-Plus
David Reitter wrote:
text(labels=123, 50, 0.5, bg=green)
I've experimented with legend(),which will make the box too wide, and
also with rect(), which doesn't know the extent of the text shown.
strwidth and strheight know!
here's a quickie - adjust to your specifications:
textBox
Hello all,
Is AIC calculated incorrectly in lmer? It appears as though it uses
AIC = -2*logLik - 2*#parms, instead of -2*LogLik + 2*#parms? Below is
output from one of many models I have tried:
Generalized linear mixed model fit using PQL
Formula: cswa ~ pcov.ess1k + (1 | year)
Data:
Dear List,
I am new to R and find it very powerful. I would like to compute the
permutational p-value for paired data using Kolmogorov-Smirnov, but
the built-in ks.test does not have this option, unlike the t.test
which has a paired=TRUE flag. Has someone written a library or a
routine that does
Up to now, I have recognized problems with gpd(..), the function from
the package evir
I think that all these functions that estimate the parameters xi, beta for
the GPD
by given threshold mu use the function optim(..) ( gpd, fitgpd, ...)
Error example:
data1 - rgpd(1000, xi= -1.5, mu=1000,
David Reitter wrote:
It's probably a beginner's question:
How do I show text in boxes?
That is, can I specify a background color for text output with text() ?
The following doesn't work as I would expect:
text(labels=123, 50, 0.5, bg=green)
I've experimented with legend(),which will
Hello all!
Consider a dataset with a grouping structure, Group (factor)
Several treatments, Treat (factor)
Some sort of yield, Yield (numeric)
Something, possibly important, measured for each group; GroupCov (numeric)
To look for fixed effects from Treat on Yield, a first attempt could be:
m1 -
Dear List
sorry to bother you R-gurus with such an unstatistical question... but
I face a problem using Crimson Editor with R 2.01.1 that I never had
using R 2.00.1.
I already posted on the Crimson Editor forum but it seems to be VERY few
R-users there
I successfully used R
Dear R users,
I've been trying to build R from sources (in Windows) using Dr. Goto's BLAS,
unsuccessfully! I've followed the instructions in Section 3.1.2-3.1.3 of R
Installation and Administration manual (but maybe I did something wrong), but I
keep receiving the following error:
-- initially I
Now I want to evaluate GroupCov as a covariate to Treat. I
can then start with either m1 or m2 as base, but what is most
correct when GroupCov has only one value for each Group?
m3 - lm(Yield ~ Treat + GroupCov + Treat:GroupCov)
gives the same fixed effects as
m4 - lme(Yield ~ Treat +
The calculation is being done in
print(data.frame(AIC = AIC(llik), BIC = BIC(llik),
logLik = c(llik),
deviance = -2*llik,
row.names = ))
where llik is defined as
llik -
Please tell us which version of R.
If this is 2.2.0 (I think it is) you need to review the tools you are
using, as mingw-runtime-3.8 is required. (It says so explicitly in that
manual, and that is the current version of that package.)
For R-patched you will also need a recent enough version
Hi,
I have the problem that for the step procedure stops due to missing
values. There are no options in Step or stepAIC to handle missing
values. Is there any way to run stepwise modelselection in R in an
automated way in this case?
Here is the last step before it stops. Hope someone knows.
On 7 Oct 2005, at 23:50, Jim Lemon wrote:
Have a look at the plotrix package, particularly textbox and
boxed.labels. I would like to know if there are any enhancements
you could suggest, as I am close to uploading the next version.
Great, that's what I needed.
thigmophobe.labels would
On Fri, 7 Oct 2005, Andreas Cordes wrote:
I have the problem that for the step procedure stops due to missing
values. There are no options in Step or stepAIC to handle missing
values. Is there any way to run stepwise modelselection in R in an
automated way in this case?
Try the hint it gives
I agree with Martin R Graph Gallery has a lot of neat stuff.
I also think there should be a on CRAN website a list of R websites. And
in a perfect world one of the search engines scope would get extended to
search them all.
Jarek
\
Hi,when I perform SAM on my array data(siggenes)I have some problems in
retrieving the separate lists of up regulated and down regulated genes.
When I write:
fold-function(x){
gruppi-split(x,controllo)
geni1-abs(mean(gruppi[[2]])-mean(gruppi[[1]]))
return(geni1)
Dear all,
I want to make some possible neighbourhoods in dataset below,
V1 - c(0,1,2,3)
V2 - c(0,1)
V3 - c(0,1,2)
V4 - c(0,1,2,3,4)
and then I have a domain which the number of each variables.
For dataset above a domain,
domains - c(3,1,2,4)
To create the neighbourhoods I choice one
Hi,
my reply just concerns the usage of AIC in mixed models and not the lmer
package.
The standard AIC is actually unconditional.
Vaida and Blanchard (2003, Proceeding 19 IWSM,101-105) discuss that a
conditional version should be more appropriate in a mixed framework.
I don't whether the
All,
I'm having a problem selecting directly from a vector. I've written ways to do
this
indirectly, but I'd rather do it directly and didn't see this in the manual.
Essentially, I have:
group.label.new
[1] 7 9 6 1 10 4 8 3 2 5
junk
[1] 1 2
group.label.new[junk 8:10]
[1] 7 9
Hi,
how about group.label.new[c(junk,8:10)]
Nolwenn
**
Nolwenn Le Meur, PhD
Fred Hutchinson Cancer Research Center
Computational Biology
1100 Fairview Ave. N., M2-B876
P.O. Box 19024
Seattle, WA 98109-1024
On Fri, 7 Oct 2005, Afshartous, David wrote:
Hi all,
I was unable to obtained a smoothed line using the loess function.
I used the following code reported in the examples of R documentation:
cars.lo - loess(dist ~ speed, cars)
Then I tried to plot both the data and the smoothed line
plot(cars)
lines(cars.lo)
but what I obtained is
On Fri, 7 Oct 2005, Dimitri Giunchi wrote:
Hi all,
I was unable to obtained a smoothed line using the loess function.I used the
following code reported in the examples of R documentation:
cars.lo - loess(dist ~ speed, cars)
Then I tried to plot both the data and the smoothed line
Hi, Christian
you must sort all data in data1, for using gpd( ...) function.
So try this following code :
library(evir)
data1 - rgpd(1000, xi= -1.5, mu=1000, beta=100)
out - gpd(sort(data1), threshold = 1060)
out
Have a good week end!
Lassana KOITA
Service Technique de l'Aviation Civile
Does anyone know how -log(x) can equal 743 but -log(x+0)=Inf? That's what
the following stream of calculations suggest:
Browse[2] -log ( 1e-323+yMat2 - yMat1 * logitShape(matrix(parsList$Xs,
nrow = numXs, ncol=numOfCurves), matrix(means, nrow = numXs,
ncol=numOfCurves, byrow=TRUE),
On Fri, 7 Oct 2005, Peter Muhlberger wrote:
Does anyone know how -log(x) can equal 743 but -log(x+0)=Inf? That's what
the following stream of calculations suggest:
Browse[2] -log ( 1e-323+yMat2 - yMat1 * logitShape(matrix(parsList$Xs,
nrow = numXs, ncol=numOfCurves), matrix(means, nrow =
I have a data file from which I need to read portions of
data but data location/quantity can change from file to file.
I wrote some code and have a working solution but it seems
wasteful to have to do it this way. Here's the contrived
incomplete code.
datalines - readLines(datafile.pathname)
Why not use a text connection?
On Fri, 7 Oct 2005, Paul Roebuck wrote:
I have a data file from which I need to read portions of
data but data location/quantity can change from file to file.
I wrote some code and have a working solution but it seems
wasteful to have to do it this way. Here's
Dear all,
In order to ease the transition from SPSS to R for some of my colleagues, I am
trying to create a function which would show the variables and their labels
(if those exist), using function label in package Hmisc.
A toy example would be this:
my.data - data.frame(age=c(24,35,28),
Adrian DUSA wrote:
Dear all,
In order to ease the transition from SPSS to R for some of my colleagues, I
am
trying to create a function which would show the variables and their labels
(if those exist), using function label in package Hmisc.
A toy example would be this:
my.data -
All,
When defining the arguments of a function, is it possible to supply a function
as
an argument? If so, how is this introduced into the function code as well?
For example, in the body of the function I have:
result = function(x)
and I'd like to supply either function.1 or
On Friday 07 October 2005 20:55, Sundar Dorai-Raj wrote:
Adrian DUSA wrote:
[...snip...]
Hi, Adrian,
You need to assign fix(dataf) to something:
my.data - data.frame(age=c(24,35,28), gender=c(Male, Female, Male))
require(Hmisc)
label(my.data$age) - Respondent's age
Adrian DUSA wrote:
On Friday 07 October 2005 20:55, Sundar Dorai-Raj wrote:
Adrian DUSA wrote:
[...snip...]
Hi, Adrian,
You need to assign fix(dataf) to something:
my.data - data.frame(age=c(24,35,28), gender=c(Male, Female, Male))
require(Hmisc)
label(my.data$age) - Respondent's age
Hello,
I noticed that the README found in /bin/windows/contrib/ATLAS indicates that
the ATLAS version is 3.4.1. According to the ATLAS sourceforge site, 3.6.0
the latest stable version.
Does anybody know if the ATLAS Rblas.dll are 3.4.1 or 3.6.0, and if they are
not the latest version, is
On Friday 07 October 2005 21:22, Sundar Dorai-Raj wrote:
[...snip...]
My guess is you want to have your function fix my.data without having to
reassign it. The answer to that question is most likely a road you do
not want to travel. Otherwise, try searching the archives for assign
reference
It's trivial -- and many R functions do this. ?outer,?sapply for example.
Once can also return a function. ?approxfun for example
Trivial example that shows how to use ... to pass in extra arguments to fun
chooseFun-function(dat=1:10,fun=mean,...)fun(dat,...)
chooseFun()
x-rnorm(100)
Hi,
I have a 7000x7000 matrix, and each element is an integer. For each element, I
want to apply the function :
wt - 0
for(q in 1:count){
wt - wt + 0.5^(q-1)
}
I get the value of 'count' from the elements in the matrix , and want to store
the corresponding 'wt' value for that element.
I also think there should be a on CRAN website a list of R websites. And
in a perfect world one of the search engines scope would get extended to
search them all.
It's quite possible to set up your own: have a look at http://rollyo.com/
Hadley
__
I suspect that loop would be the fastest since it would have the least
memory overhead. You do not want to have too many internal copies of
7000x7000 matrices.
Jarek
\
Jarek Tuszynski, PhD. o / \
Science
Well, since Sum(i=1 to i-n) =n*(n+1)/2, your loop simply gives
1/4*count*(count-1).
So if your matrix is A, A*(A-1)/4 is about the quickest way to get your
answer I think.
-- Bert Gunter
Genentech Non-Clinical Statistics
South San Francisco, CA
The business of the statistician is to catalyze
Folks,
I've run into trouble while writing functions that I hope will create
and modify a dataframe or two. To that end I've written a toy function
that simply sets a couple of variables (well, tries but fails).
Searching the archives, Thomas Lumley recently explained the -
operator, showing
If A is the matrix the answer is 2*(1 - 2^(-A)), which took about 10secs
for an example of your size.
From \sum_{i=1}^n x^{1-i} = (1-x^{-n})/(1-x), E OE.
On Fri, 7 Oct 2005, Tim Smith wrote:
I have a 7000x7000 matrix, and each element is an integer. For each
element, I want to apply the
On Fri, 7 Oct 2005, Prof Brian Ripley wrote:
If A is the matrix the answer is 2*(1 - 2^(-A)), which took about 10secs
for an example of your size.
From \sum_{i=1}^n x^{1-i} = (1-x^{-n})/(1-x), E OE.
NB: I have assumed n = 1 here, as people nornally do when using 1:count.
On Fri, 7 Oct
Oops -- quite right, Brian. Interesting, as I stared at it for some time.
Sigh... now what **WAS** I looking for in the refrigerator?
-- Bert Gunter
Genentech Non-Clinical Statistics
South San Francisco, CA
-Original Message-
From: Prof Brian Ripley [mailto:[EMAIL PROTECTED]
Hi Thomas: Thanks! Yes, the function
(yMat2[5,9]-yMat1[5,9]*logitShape()[5,9]) appears to be producing a value of
-1.102216e-16 rather than 0. I would have thought it would approach 0 from
above given that all input values are at or above zero, but evidently not.
The max function won't do the
If there weren't an analytic solution to your problem,
then you could build a vector of the answers from 1
to the maximum in the matrix. Call that 'wtvec'. Then:
ans - array(NA, dim(A), dimnames(A))
ans[] - wtvec[as.vector(A)]
should get you what you want.
Patrick Burns
[EMAIL PROTECTED]
+44
On Fri, 7 Oct 2005, Patrick Burns wrote:
If there weren't an analytic solution to your problem,
then you could build a vector of the answers from 1
to the maximum in the matrix. Call that 'wtvec'. Then:
ans - array(NA, dim(A), dimnames(A))
ans[] - wtvec[as.vector(A)]
should get you what
Hello,
Does anyone know if the cor function, when method = spearman, returns a
correlation coefficient corrected for any ties in the ranks of the data? I
have data with quite a few ties and am thinking that I should use a calculation
of the coefficient corrected for ties, but before I try and
The sum in the loop is simply: 2 - (0.5)^count.
So you don't need this loop. As count gets large, the sum approaches 2.
Ravi.
-Original Message-
From: [EMAIL PROTECTED] [mailto:r-help-
[EMAIL PROTECTED] On Behalf Of Patrick Burns
Sent: Friday, October 07, 2005 4:29 PM
To: Tim
Rather than adding 1e-15 to all numbers, I suggest you simply make
that the floor. (Or use .Machine$double.eps or 2*.Machine$double.eps in
place of 1e-15.)
Another alternative that may or may not apply in your case is to
develop an asymptotic expansion for the
Hello:
I have a matrix 'dat' with 2 columns.
I have the following code:
for (i in 1:nrows(dat))
{
if (dat[i,1] dat[i,2])
{
dat[i,2]-0
}
else
{
dat[i,2]-1
}
Is there a way to accomplish this without the for loop?
Thank you.
-Dhiren
Dear R users
I am trying to write an R function to solve for a,b,c in the following
system of equations, given any value of x1, x2 and x3:
b^2 + 6*b*a + 2*c^2 + 15*a^2 = x1
2*c*(b^2 + 24*b*a + 105*a^2 + 2) = x2
24*(b*a + c^2*(1 + b^2 + 28*b*a) + a*(12 + 48 *b*a + 141*c^2 + 225*a^2))
=x3
Could
On Fri, 07 Oct 2005 18:01:41 -0400 Dhiren DSouza wrote:
Hello:
I have a matrix 'dat' with 2 columns.
I have the following code:
for (i in 1:nrows(dat))
{
if (dat[i,1] dat[i,2])
{
dat[i,2]-0
}
else
{
dat[i,2]-1
}
Is there a way to accomplish
Looking at what objects exist after the call to myFunk() should give you
a clue as to what happened:
remove(list=objects())
myFunk-function(a,b,foo,bar) {foo-a+b; bar-a*b;}
x-0; y-0;
myFunk(4,5,x,y)
x
[1] 0
y
[1] 0
objects()
[1] barfoomyFunk x y
bar
[1] 20
foo
[1]
Hi,
Tried to install AnalyzeFMRI on Linux (FC-4) and got the following:
install.packages(AnalyzeFMRI)
trying URL
'http://rh-mirror.linux.iastate.edu/CRAN/src/contrib/AnalyzeFMRI_1.1-4.tar.gz'
Content type 'application/x-gzip' length 308066 bytes
opened URL
Standard answers:
1) Is your version of R up to date?
2) Is the package version up to date?
3) Contact the package maintainer.
BTW, I believe that list etiquette is to provide your real name.
-- Bert Gunter
Genentech Non-Clinical Statistics
South San Francisco, CA
The business of the
Hi,
Many thanks! I was able to install other packages so I will contact the package
maintainer if I can figure out who he is. Any clues?
I use this Yahoo! e-mail address as a junk e-mail address since the archives
contain the complete e-mail address (no anti-spam measures taken) so I do not
Hi,
I am trying to convert the following NLMIXED code to NLME, but am
running into problems concerning 'Singularity in backsolve'. As I am new
to R/S-Plus, I thought I may be missing something in the NLME code.
NLMIXED
***
proc nlmixed data=kidney.kidney;
parms delta=0.03 gamma=1.1
Tim Smith wrote:
Hi,
I have a 7000x7000 matrix, and each element is an integer. For each element,
I want to apply the function :
wt - 0
for(q in 1:count){
wt - wt + 0.5^(q-1)
}
This is not a function! Maybe you want
helper - function(count) sum(0.5^((1:count)-1))
I get the
72 matches
Mail list logo